Attention decomposes into low-rank routing and symmetric filtering; disentangled S-D attention reveals a spectral cascade allowing early-layer linearization at under 5% perplexity cost.
Attention is all you need,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
The Routing and Filtering Structure of Attention
Attention decomposes into low-rank routing and symmetric filtering; disentangled S-D attention reveals a spectral cascade allowing early-layer linearization at under 5% perplexity cost.